Long-term cohort study of patients presenting with hypercapnic respiratory failure.

ABSTRACT

OBJECTIVE: We sought to describe the long-term prognosis for a population-based cohort of people with hypercapnic respiratory failure (HRF) and the associations between underlying diagnoses and the risks of death and rehospitalisation. METHODS: We performed a historical cohort study of all persons with HRF in the Liverpool local government area in New South Wales, Australia, in the 3-year period from 2013 to 2015. Cohort members were identified using arterial blood gas results from Liverpool Hospital demonstrating pH ≤7.45 and PaCO2 >45 mm Hg within 24 hours of presentation. Linked health data were obtained from statewide registries with a minimum follow-up period of 6 years. The primary outcomes were time to death from any cause and the standardised mortality ratio (SMR) which compares the observed to the expected number of deaths in the same population. Secondary outcomes were time to rehospitalisation and the associations between death and/or hospitalisation and underlying diagnoses. RESULTS: The cohort comprised 590 adults aged between 15 and 101 years. Overall, 415 (70.3%) participants died in the follow-up period. Among those who survived the index admission, the probability of survival at 1, 3 and 5 years was 81%, 59% and 45%, respectively. The overall SMR was 9.2 (95% CI 7.6 to 11.0), indicating a near 10-fold risk of death than otherwise expected for age. Most (91%) survivors experienced rehospitalisation, with median (IQR) time to readmission of 3.9 (1.2-10.6) months. Congestive cardiac failure and neuromuscular disease were associated with an increased risk of death, whereas chronic obstructive pulmonary disease and sleep disordered breathing increased the risk of rehospitalisation. CONCLUSIONS: HRF is associated with poor survival and high risk of rehospitalisation in the 5 years following an index event. The underlying disease appears to have some influence on overall survival and subsequent hospitalisations.